Using OpenCV, NumPy and several classical filters, perform live edge detection from a webcam source.
Usage:
python3 detect_edges.py
Customiseable options:
filters_to_use is a customiseable dictionary of filters. Built-in for easy use are the the classical Prewitt, Sobel, Robinson, and Kirsch edge-detecting filters. In the dictionary, ensure the key argument is a string indicating the kernel's name, and the value the numpy array.
height, width are options to change the number of copies of the webcam to see in each row and column. Any grid can be made, and if a grid is made that holds more elements than filters tested - remaining space will be zero-d out. In the picture above, I set width=4, height=3.